Summary
- Google AI Studio introduced vibe coding, which allows you to create AI-powered apps using Gemini to automatically integrate video and image APIs.
- The platform allows the description of multimodal apps and creates them based on text prompts, without the need for programming knowledge.
- AI Studio offers a free quota, with the option to add an API key for continued use after the limit is exhausted.
Google added a new vibe coding experience to AI Studio, the company’s platform with AI models still under development, but available for testing by the public. The goal, as usual, is to allow anyone (even without programming knowledge) to create a functional AI app, using text prompts.
The tool uses the latest Gemini templates to do the heavy lifting. The user describes the app they want, just as they would ask for an image to be developed, and the AI automatically connects the models and APIs needed to make it work.
The new vibe coding platform joins Opal — which arrived in Brazil this month — in the search giant’s portfolio. In AI Studio, aimed at developers, the idea is for applications to integrate company APIs and even Gemini prototypes before a large-scale application.
How does it work?
According to Google, the user can describe the multimodal application they want. For example, when asking for an app that “generates videos from a script” or “creates a photo editing tool”, AI Studio understands the command and automatically connects the necessary models, such as Veo or Nano Banana, respectively.
However, this definition can also be manual. Below the text box, the platform makes all models and APIs available for the user to select.
To help, the platform also redesigned the app gallery, transforming it into a visual library of projects. Users can explore ideas, learn from the initial code, and remix apps for their own creations.
In addition to the initial creation, vibe coding brings a new Annotation Mode. With it, the user can change the app visually, without having to edit the source code directly. Just highlight an element in the interface (like a button or a card) and instruct Gemini with commands like “Change the color of this button to blue”.
Free usage limit
Access to AI Studio operates with a free usage quota. Google informs that, if the user runs out of the limit, they can add their own API key (associated with a paid Google Cloud project) to “continue vibe coding without interruption”.
According to the company, the goal is to “reduce the barrier between a great idea and a working app with Gemini, so anyone can build with AI.” The idea is bold at the moment: even the person responsible for coining the term “vibe coding”, former OpenAI Andrej Karpathy, revealed that the method is still limited when it comes to larger projects.
Source: https://tecnoblog.net/noticias/vibe-coding-chega-ao-google-ai-studio-veja-como-criar-apps-com-ia/
